mb706

Use `missings.numerics`, `missings.factors`, `missings.ordered`, `missings.other` instead of `missings` property (and use `missings` as a shorthand of all four of these, *or* as a subset, depending on `dataformat` and `affect`). Many...

since it doesn't compare number of occurrences

Probably can simplify some internals by handling a `dataformat = ""` CPO like a CPO that has `affect.type = ""` always set. This would also simplify potential column skips (skip...

...because of mlr-org/mlr#2099. If mlr-org/mlr#2108 gets solved, this would need to be solved using the `"missings.predict"` property.

e.g. model.matrix

Skipping with zero rows might be dangerous if the resulting retrafo is expected to do something useful, so it is better to let the method fail in that case. Skipping...

e.g. ```R par.set = pSS(do.impute: logical), properties.adding = function(do.impute) { if (missing(do.impute)) "sometimes.missing" else if (do.impute) "missings" else character(0) } ``` Where only unexported parameters are given and hence `missing()`...

We are in the process of overhauling the `paradox` package, on which your `NADIA` package depends: At some point in the near future, we are going to merge [this PR](https://github.com/mlr-org/paradox/pull/350)....